Proceedings of the General Track: 2004 USENIX Annual Technical Conference

Recent work on Internet measurement and overlay networks has shown that redundant paths are common between pairs of hosts and that one can often achieve better end-to-end performance by adaptively choosing an alternate path [8, 28]. In this paper, we propose an end-to-end transport layer protocol, mTCP, which can aggregate the available bandwidth of those redundant paths in parallel. By striping one flow's packets across multiple paths, mTCP can not only obtain higher end-to-end throughput but also become more robust under path failures. When some paths fail, mTCP can continue sending packets on other living paths and the recovery process normally takes only a few seconds. Because mTCP could obtain an unfair share of bandwidth under shared congestion, we integrate a shared congestion detection mechanism into our system. It allows us to dynamically detect and suppress paths with shared congestion so as to alleviate the aggressiveness problem. mTCP can also passively monitor the performance of several paths in parallel and discover better paths than the path provided by the underlying routing infrastructure. We also propose a heuristic to find disjoint paths between pairs of nodes using traceroute. We have implemented our system on top of overlay networks and evaluated it in both Planet-Lab and Emulab.

[1]  V. Paxson.  End-to-end routing behavior in the internet , 2006, CCRV.

[2]  Anees Shaikh,et al.  A measurement-based analysis of multihoming , 2003, SIGCOMM '03.

[3]  John G. Apostolopoulos,et al.  On multiple description streaming with content delivery networks , 2002, Proceedings.Twenty-First Annual Joint Conference of the IEEE Computer and Communications Societies.

[4]  Azer Bestavros,et al.  Robust identification of shared losses using end-to-end unicast probes , 2000, Proceedings 2000 International Conference on Network Protocols.

[5]  Ratul Mahajan,et al.  Measuring ISP topologies with Rocketfuel , 2004, IEEE/ACM Transactions on Networking.

[6]  Sally Floyd,et al.  TCP Selective Acknowledgement Options , 1996 .

[7]  Brian D. Noble,et al.  The end-to-end performance effects of parallel TCP sockets on a lossy wide-area network , 2002, Proceedings 16th International Parallel and Distributed Processing Symposium.

[8]  Anindo Banerjea.  Simulation Study of the Capacity Effects of Dispersity Routing for Fault Tolerant Realtime Channels , 1996, SIGCOMM.

[9]  Akihiro Nakao,et al.  A routing underlay for overlay networks , 2003, SIGCOMM '03.

[10]  Yin Zhang,et al.  On the constancy of internet path properties , 2001, IMW '01.

[11]  Bernard Chazelle,et al.  The discrepancy method - randomness and complexity , 2000 .

[12]  Robert L. Grossman,et al.  PSockets: The Case for Application-level Network Striping for Data Intensive Applications using High Speed Wide Area Networks , 2000, ACM/IEEE SC 2000 Conference (SC'00).

[13]  Raghupathy Sivakumar,et al.  A Transport Layer Approach for Achieving Aggregate Bandwidths on Multi-Homed Mobile Hosts , 2002, MobiCom '02.

[14]  Yin Zhang,et al.  On the characteristics and origins of internet flow rates , 2002, SIGCOMM '02.

[15]  Ossama Younis,et al.  On efficient on-line grouping of flows with shared bottlenecks at loaded servers , 2002, 10th IEEE International Conference on Network Protocols, 2002. Proceedings..

[16]  Anja Feldmann,et al.  Deriving traffic demands for operational IP networks: methodology and experience , 2000, SIGCOMM.

[17]  Robin Kravets,et al.  Transport level mechanisms for bandwidth aggregation on mobile hosts , 2001, Proceedings Ninth International Conference on Network Protocols. ICNP 2001.

[18]  David E. Culler,et al.  A blueprint for introducing disruptive technology into the Internet , 2003, CCRV.

[19]  Jonathan M. Smith,et al.  Striping within the network subsystem , 1995, IEEE Netw..

[20]  Brian Tierney,et al.  Applied techniques for high bandwidth data transfers across wide area networks , 2001 .

[21]  A. Banerjea.  Simulation Study of the Capacity Effects of Dispersity Routing for Fault Tolerant Realtime Channels , 1996, SIGCOMM.

[22]  George Varghese,et al.  A Reliable and Scalable Striping Protocol , 1996, SIGCOMM.

[23]  Mark Allman,et al.  An Application-Level solution to TCP''s Satellite Inefficiencies , 1996 .

[24]  Avideh Zakhor,et al.  Path diversity with forward error correction (PDF) system for packet switched networks , 2003, IEEE INFOCOM 2003. Twenty-second Annual Joint Conference of the IEEE Computer and Communications Societies (IEEE Cat. No.03CH37428).

[25]  Donald F. Towsley,et al.  Detecting shared congestion of flows via end-to-end measurement , 2000, SIGMETRICS '00.

[26]  Raghupathy Sivakumar,et al.  pTCP: an end-to-end transport layer protocol for striped connections , 2002, 10th IEEE International Conference on Network Protocols, 2002. Proceedings..

[27]  J. Duncanson,et al.  Inverse multiplexing , 1994, IEEE Communications Magazine.

[28]  Dina Katabi,et al.  An Information Theoretic Approach for Shared Bottleneck Inference Based on End-to-end Measurements , 1999 .

[29]  Anees Shaikh,et al.  An empirical evaluation of wide-area internet bottlenecks , 2003, SIGMETRICS '03.

[30]  Alex C. Snoeren,et al.  Adaptive inverse multiplexing for wide-area wireless networks , 1999, Seamless Interconnection for Universal Services. Global Telecommunications Conference. GLOBECOM'99. (Cat. No.99CH37042).

[31]  Robert Tappan Morris,et al.  Resilient overlay networks , 2001, SOSP.

[32]  RubensteinDan,et al.  Detecting shared congestion of flows via end-to-end measurement , 2002 .

[33]  Stefan Savage,et al.  The end-to-end effects of Internet path selection , 1999, SIGCOMM '99.

[34]  Bernd Girod,et al.  Real-time voice communication over the internet using packet path diversity , 2001, MULTIMEDIA '01.